首页> 外文学位 >Collaboration in language testing: Investigating the validity of a new rating procedure for EFL writing assessment in Korea.
【24h】

Collaboration in language testing: Investigating the validity of a new rating procedure for EFL writing assessment in Korea.

机译:在语言测试方面的合作:调查韩国EFL写作评估新评级程序的有效性。

获取原文
获取原文并翻译 | 示例

摘要

This study seeks an innovative testing system in which trained students participate in a writing test as a complementary rater, specifically in the EFL context. The study has three research goals: (1) to investigate the psychometric reliability of self- and peer-ratings in a second language (L2) essay-writing test in Korea, (2) to explore students' rating process, and (3) to understand student's perception of rating tasks. The rationale of this research is that the Korean EFL students whose major is English education should conduct the rating tasks that they will encounter in their future career. Each of 104 Korean EFL learners at the intermediate level sequentially rated three sample essays, his/her own essay, and his/her partner's essay after he/she was provided with the same benchmark as two experts would refer to. Each essay was rated with a 5-scale in terms of 4 writing features (Organization, Content, Language Use, and Holistic feature). After students finished the rating tasks, they filled out a questionnaire. The students' ratings then were compared to expert ratings, using Multi-facet Rasch model. The quantitative research was accompanied by the qualitative study. Eleven students participated in semi-structured interviews. For the internal and external validity of the research, the researcher took use of three types of triangulations: data triangulation, methodology triangulation, and investigator triangulation. Results showed that peer raters were so lenient that they showed unpredictable rating patterns. From the qualitative analyses, students actively utilized a variety of rating strategies. In addition, they showed the positive attitudes toward the rating tasks. Most interviewees, however, viewed that they were not able to be an expert rater because of their linguistic incompetence. It was concluded that the collaborative rating procedure fosters students to activate high-mental function and a social skills, raising students' critical awareness of their social context.
机译:本研究寻求一种创新的考试系统,在该系统中,受过训练的学生可以作为补充评分者参加写作考试,特别是在EFL语境中。这项研究的三个研究目标是:(1)在韩国以第二语言(L2)作文写作测试来调查自我和同伴评级的心理计量学可靠性;(2)探索学生的评级过程;以及(3)了解学生对评分任务的看法。这项研究的基本原理是,以英语教育为专业的韩国EFL学生应该执行他们未来职业中将会遇到的评分任务。向104位韩国中级英语学习者中的每位学习者依次授予三份示例性论文,他/她自己的论文以及他/她的伴侣的论文,然后给他/她提供与两名专家相同的基准。根据4种写作功能(组织,内容,语言使用和整体功能),每篇论文的等级分为5级。学生完成评分任务后,他们填写了问卷。然后,使用多方面Rasch模型将学生的评分与专家评分进行比较。定量研究伴随着定性研究。 11名学生参加了半结构式访谈。为了研究的内部和外部有效性,研究人员使用了三种类型的三角剖分:数据三角剖分,方法三角剖分和调查者三角剖分。结果表明,同行评等人宽容,以至于显示出不可预测的评等模式。通过定性分析,学生们积极运用了各种评分策略。此外,他们对评级工作表现出积极态度。但是,大多数受访者认为,由于他们的语言能力不强,他们无法成为专家评级者。结论是,协作评估程序可以培养学生激活高智商的功能和社交技能,从而提高学生对其社交环境的批判意识。

著录项

  • 作者

    Lee, Ho.;

  • 作者单位

    University of Illinois at Urbana-Champaign.;

  • 授予单位 University of Illinois at Urbana-Champaign.;
  • 学科 Education Language and Literature.; Education Tests and Measurements.; Education Educational Psychology.
  • 学位 Ph.D.
  • 年度 2005
  • 页码 228 p.
  • 总页数 228
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 教育;教育心理学;
  • 关键词

  • 入库时间 2022-08-17 11:42:13

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号